Also, I'm not finding a lot of information about loss of glycosylation sites as a general occurrence in serial passage through cell culture. Can you provide some links about that? Did it occur in any of the furin cleavage research I linked previously? It's my understanding that at least on the S protein, glycosylation plays an important role in the virus's entry into cells, and not only in evading the immune system, which implies it would be conserved even in culture. Furthermore, the selective pressure of an immune system could be restored by using whole live animals. Transgenic mice expressing the human ACE2 receptor are currently being used to study SARS-CoV-2; perhaps they were also used in its development?
Regarding the hACE2 transgenic mouse, you are talking about an animal model that was developed in 4 laboratories between 2002 and 2006. This animal model was not commonly used around the world at all. I do believe that it was in use at the WIV, based on a statement in this episode of This Week in Virology (listen around 1 hour in): https://www.microbe.tv/twiv/twiv-591/
The thing is, these mice are not exactly a great virus production system, at least as far as I know. They are designed to be permissive to infection and to facilitate the study of SARS-CoV-1, and they develop something that resembles but is of course not identical to the human disease. It would be extremely weird to passage a virus in them for the purposes of creating it. The yields would probably be low. The virus would become increasingly mouse-adapted, because aside from expressing hACE2, they're still largely regular lab mice. SARS-CoV viruses do develop mouse-specific adaptations, and mouse adaptation makes the viruses easier to study in these models. If this were a mouse-adapted virus, that would have been pretty apparent early on.
But this is all a deep dive for a question on which there is currently no supporting evidence, as well as contrary epidemiological evidence. It remains that there is no epidemiological evidence of emergence of this virus within the actual city of Wuhan prior to the appearance of cases in more rural areas of Hubei, where there are bats and people who interact with them. There are active questions to be asked on all fronts of this investigation, but to pursue the concept of this least likely possibility above all others, even in the face of incompatible evidence, doesn't seem like a great use of resources. I can come up with all manner of "possible" origin stories for this virus. Possible and likely are different concepts, and I'm going to be focusing on "likely."
In the event that investigations uncover something that overturns that evidence, priorities should be changed, but this seems like a lot of mental and experimental acrobatics to try to make it possible for the virus to have come from WIV when we have better reason to believe it came from rural areas of Hubei province.
Until there is new evidence--and I mean something really genuine that can be scientifically assessed, not conjecture or speculation--I can't change my priors on this. The evidence available suggests that a natural origin is most likely. Nothing new has been presented that makes another option more likely. I'll wait for evidence. If it comes, I'll report on it. Right now there's nothing more to be said on this topic.
Yuri Deigin's writeup on the lab leak hypothesis links to five different papers from 2006 to 2019 that involve messing with furin cleavage sites of coronaviruses, including SARS-CoV. I don't have the background to understand this research in depth, but from my layman's perspective it seems strange to say that there is no evidence of anyone experimenting in this area. The papers are:
I don't believe I have ever said there is an absence of evidence of people modifying the spike protein of coronaviruses in model systems, which is the only thing being done in any of these papers. No viruses are made here, but even if they were, none of these would represent evidence that SARS-CoV-2 was engineered in any fashion, nor would they provide any support for the idea that this alleged engineering took place at WIV.
To address one by one, the links you share here are:
1) A study where SARS-CoV-1 Spike protein constructs--not virus--are edited in plasmids and then studied as plasmids. Nothing here is infectious, and nothing here is an infection-competent virus with similarity to RaTG13 or SARS-CoV-2. This is a study conducted in Montana, not the WIV.
2) Another study where an artificial, incompetent virus system is used as a model to study changes to the SARS-CoV-1 S protein. This one uses a pseudotyped (ie surface proteins are different) lentivirus system (a retrovirus model that is frequently used to do this type of work). Again, nothing here is a human infectious agent and the constructs are not similar to RaTG13 or SARS-CoV-2. This is a study conducted in Tokyo, not the WIV.
3) This is a study that again uses plasmid constructs of the SARS-CoV-1 S protein which are mutated using typical techniques for modifying bacterial plasmid constructs. Nothing here is an infectious agent. Nothing here is similar to the full-length genome of RaTG13 or SARS-CoV-2. This is a study conducted in the Netherlands, not the WIV.
4) This paper mixes plasmid work with pseudotyped virus work, again trying to modify only the S protein. Nothing here is infectious to humans and nothing is a competent virus resembling RaTG13 or SARS-CoV-2. This is a study from New York, not the WIV.
5) Finally we have a paper that made some virus! Specifically, this paper made recombinant
avian infectious bronchitis virus (IBV), a pathogen that infects domestic chickens. This virus is a coronavirus from the genus Gammacoronavirus, which is far removed from the Betacoronavirus genus, and it only infects birds. The focus of this paper is on the cleavage of the spike protein, and how that impacts the virus. To investigate this, they mutate the spike protein and see what happens with the virus. There is nothing here even vaguely related to SARS-CoV-2 (or 1), RaTG13, or at all infectious to humans. This work was conducted in China--specifically in Beijing, which is 716 miles from Wuhan. That's about the distance from New York City to Irmo, SC, which is approximately in the middle of SC.
My point here is to illustrate the incredibly circumstantial and disconnected nature of such "evidence." What these demonstrate is that various labs in the world are capable of recombinant genetics. That is not new information, of course.
I think Yuri Deigin has done some interesting contrarian work, but on the whole he's making an argument out of extremely circumstantial points merely to suggest that it might, perhaps, be possible for SARS-CoV-2 to have been engineered. This is so remote and unlikely that I really think he's wasting his time. I believe he has demonstrated that technology exists to create virus sequences of certain types, but this doesn't demonstrate that anyone in the world has the technology to predict which sequences will create pandemic-potential viruses and which will create useless broken assemblies of nucleotides. The SARS-CoV-2 genome contains nothing that appears to be a targeted change; it contains many random changes, with clearly inefficient mutational insertions instead. It is not optimized for passage in humans (it continues to evolve new variants) or in other animals (again, we see variants appearing in animals like minks). Nothing about this suggests an intelligent design origin. Nothing Deigin presents demonstrates anything about SARS-CoV-2 in particular, which continues to be a generalized flaw of the "engineered virus" camp. Extraordinary claims require, at the very least, some direct evidence.
I'm interested in real testing of hypotheses, not speculation based on the work of other laboratories. The Andersen paper, which is the foundational work that Deigin has been trying to argue with, suggests several testable hypotheses and then proceeds to test them in order to make a meaningful case that the virus is unlikely to have been engineered. Please note that this is not an argument that it is "impossible" for it to have been engineered, but it is indeed extremely unlikely when compared to other completely reasonable options.
To establish a laboratory origin for SARS-CoV-2, Deigin also needs to develop and test specific hypotheses. Right now he seems to be in the camp of accidental release, so he's not trying to propose that the WIV engineered the virus and then intentionally released it. If release is accidental, then Deigin would need to demonstrate that there was a cluster of SARS-CoV-2 infection and COVID-19 disease centered on or connected to the WIV before there was disease anywhere else. Right now there is no such evidence. If such evidence were to be found, it would support an accidental laboratory release of *something,* but not necessarily an engineered virus.
To establish engineering, Deigin would need to find some kind of physical evidence of the specific recombineering he alleges at a facility in Wuhan. I am assuming that he would start with the WIV because that's what everyone is looking at. So far, Phase One of the WHO investigation has identified no evidence of such recombineering at the WIV. No one has identified any evidence of it, in fact. Physical evidence would include, but not be limited to:
-Primer sets that could be used for site-directed mutagenesis
-DNA synthesis orders that contain sequences similar to SARS-CoV-2; I do not believe that the WIV possesses in-house DNA synthesis capabilities (most labs do not), but I would be interested to learn if they do. They would have needed to order synthesized sequences from third-party companies and this would have left a paper trail. The investigation should look into this, but I do not expect anything to be found.
-Samples representing intermediate constructs on the path to the construction of SARS-CoV-2, resident at the WIV.
-Samples from experimental animals demonstrating infection with a similar virus, an intermediate virus, or something
So far there is nothing of the sort in evidence. This is what I mean when I say "there is no evidence."
As a side point, I really don't understand Deigin's obsession with the furin cleavage site. Plenty of coronaviruses have furin cleavage sites. It really isn't too surprising to see one emerge in a poorly-documented species of the virus. Kristian G Andersen takes down several of the arguments that Deigin makes about the furin cleavage site's appearance in a thread here: https://twitter.com/K_G_Andersen/status/1391507230848032772
Anyway, I don't make much of Deigin's arguments. Largely, they are conjecture layered on speculation, with the word "possible" carrying a lot of water for the overall scientific validity of what he tries to say. A lot of things are possible. Show me what is probable, or better yet, show me evidence of what actually happened.
Also, I'm not finding a lot of information about loss of glycosylation sites as a general occurrence in serial passage through cell culture. Can you provide some links about that? Did it occur in any of the furin cleavage research I linked previously? It's my understanding that at least on the S protein, glycosylation plays an important role in the virus's entry into cells, and not only in evading the immune system, which implies it would be conserved even in culture. Furthermore, the selective pressure of an immune system could be restored by using whole live animals. Transgenic mice expressing the human ACE2 receptor are currently being used to study SARS-CoV-2; perhaps they were also used in its development?
Serial passage experimentation showing loss of a glycosylation site is demonstrated here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3815150/
Additionally, the emergence of O-linked glycosylation sites in the presence of immune system pressure is discussed here:
https://academic.oup.com/glycob/article/28/7/443/4951691
The fitness disadvantages of introduced glycosylation sites in a related coronavirus spike are discussed here:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5121417/
Regarding the hACE2 transgenic mouse, you are talking about an animal model that was developed in 4 laboratories between 2002 and 2006. This animal model was not commonly used around the world at all. I do believe that it was in use at the WIV, based on a statement in this episode of This Week in Virology (listen around 1 hour in): https://www.microbe.tv/twiv/twiv-591/
The thing is, these mice are not exactly a great virus production system, at least as far as I know. They are designed to be permissive to infection and to facilitate the study of SARS-CoV-1, and they develop something that resembles but is of course not identical to the human disease. It would be extremely weird to passage a virus in them for the purposes of creating it. The yields would probably be low. The virus would become increasingly mouse-adapted, because aside from expressing hACE2, they're still largely regular lab mice. SARS-CoV viruses do develop mouse-specific adaptations, and mouse adaptation makes the viruses easier to study in these models. If this were a mouse-adapted virus, that would have been pretty apparent early on.
But this is all a deep dive for a question on which there is currently no supporting evidence, as well as contrary epidemiological evidence. It remains that there is no epidemiological evidence of emergence of this virus within the actual city of Wuhan prior to the appearance of cases in more rural areas of Hubei, where there are bats and people who interact with them. There are active questions to be asked on all fronts of this investigation, but to pursue the concept of this least likely possibility above all others, even in the face of incompatible evidence, doesn't seem like a great use of resources. I can come up with all manner of "possible" origin stories for this virus. Possible and likely are different concepts, and I'm going to be focusing on "likely."
In the event that investigations uncover something that overturns that evidence, priorities should be changed, but this seems like a lot of mental and experimental acrobatics to try to make it possible for the virus to have come from WIV when we have better reason to believe it came from rural areas of Hubei province.
Until there is new evidence--and I mean something really genuine that can be scientifically assessed, not conjecture or speculation--I can't change my priors on this. The evidence available suggests that a natural origin is most likely. Nothing new has been presented that makes another option more likely. I'll wait for evidence. If it comes, I'll report on it. Right now there's nothing more to be said on this topic.
Yuri Deigin's writeup on the lab leak hypothesis links to five different papers from 2006 to 2019 that involve messing with furin cleavage sites of coronaviruses, including SARS-CoV. I don't have the background to understand this research in depth, but from my layman's perspective it seems strange to say that there is no evidence of anyone experimenting in this area. The papers are:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7111780/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2583654/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2519682/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2660061/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6832359/
I don't believe I have ever said there is an absence of evidence of people modifying the spike protein of coronaviruses in model systems, which is the only thing being done in any of these papers. No viruses are made here, but even if they were, none of these would represent evidence that SARS-CoV-2 was engineered in any fashion, nor would they provide any support for the idea that this alleged engineering took place at WIV.
To address one by one, the links you share here are:
1) A study where SARS-CoV-1 Spike protein constructs--not virus--are edited in plasmids and then studied as plasmids. Nothing here is infectious, and nothing here is an infection-competent virus with similarity to RaTG13 or SARS-CoV-2. This is a study conducted in Montana, not the WIV.
2) Another study where an artificial, incompetent virus system is used as a model to study changes to the SARS-CoV-1 S protein. This one uses a pseudotyped (ie surface proteins are different) lentivirus system (a retrovirus model that is frequently used to do this type of work). Again, nothing here is a human infectious agent and the constructs are not similar to RaTG13 or SARS-CoV-2. This is a study conducted in Tokyo, not the WIV.
3) This is a study that again uses plasmid constructs of the SARS-CoV-1 S protein which are mutated using typical techniques for modifying bacterial plasmid constructs. Nothing here is an infectious agent. Nothing here is similar to the full-length genome of RaTG13 or SARS-CoV-2. This is a study conducted in the Netherlands, not the WIV.
4) This paper mixes plasmid work with pseudotyped virus work, again trying to modify only the S protein. Nothing here is infectious to humans and nothing is a competent virus resembling RaTG13 or SARS-CoV-2. This is a study from New York, not the WIV.
5) Finally we have a paper that made some virus! Specifically, this paper made recombinant
avian infectious bronchitis virus (IBV), a pathogen that infects domestic chickens. This virus is a coronavirus from the genus Gammacoronavirus, which is far removed from the Betacoronavirus genus, and it only infects birds. The focus of this paper is on the cleavage of the spike protein, and how that impacts the virus. To investigate this, they mutate the spike protein and see what happens with the virus. There is nothing here even vaguely related to SARS-CoV-2 (or 1), RaTG13, or at all infectious to humans. This work was conducted in China--specifically in Beijing, which is 716 miles from Wuhan. That's about the distance from New York City to Irmo, SC, which is approximately in the middle of SC.
My point here is to illustrate the incredibly circumstantial and disconnected nature of such "evidence." What these demonstrate is that various labs in the world are capable of recombinant genetics. That is not new information, of course.
I think Yuri Deigin has done some interesting contrarian work, but on the whole he's making an argument out of extremely circumstantial points merely to suggest that it might, perhaps, be possible for SARS-CoV-2 to have been engineered. This is so remote and unlikely that I really think he's wasting his time. I believe he has demonstrated that technology exists to create virus sequences of certain types, but this doesn't demonstrate that anyone in the world has the technology to predict which sequences will create pandemic-potential viruses and which will create useless broken assemblies of nucleotides. The SARS-CoV-2 genome contains nothing that appears to be a targeted change; it contains many random changes, with clearly inefficient mutational insertions instead. It is not optimized for passage in humans (it continues to evolve new variants) or in other animals (again, we see variants appearing in animals like minks). Nothing about this suggests an intelligent design origin. Nothing Deigin presents demonstrates anything about SARS-CoV-2 in particular, which continues to be a generalized flaw of the "engineered virus" camp. Extraordinary claims require, at the very least, some direct evidence.
I'm interested in real testing of hypotheses, not speculation based on the work of other laboratories. The Andersen paper, which is the foundational work that Deigin has been trying to argue with, suggests several testable hypotheses and then proceeds to test them in order to make a meaningful case that the virus is unlikely to have been engineered. Please note that this is not an argument that it is "impossible" for it to have been engineered, but it is indeed extremely unlikely when compared to other completely reasonable options.
To establish a laboratory origin for SARS-CoV-2, Deigin also needs to develop and test specific hypotheses. Right now he seems to be in the camp of accidental release, so he's not trying to propose that the WIV engineered the virus and then intentionally released it. If release is accidental, then Deigin would need to demonstrate that there was a cluster of SARS-CoV-2 infection and COVID-19 disease centered on or connected to the WIV before there was disease anywhere else. Right now there is no such evidence. If such evidence were to be found, it would support an accidental laboratory release of *something,* but not necessarily an engineered virus.
To establish engineering, Deigin would need to find some kind of physical evidence of the specific recombineering he alleges at a facility in Wuhan. I am assuming that he would start with the WIV because that's what everyone is looking at. So far, Phase One of the WHO investigation has identified no evidence of such recombineering at the WIV. No one has identified any evidence of it, in fact. Physical evidence would include, but not be limited to:
-Primer sets that could be used for site-directed mutagenesis
-DNA synthesis orders that contain sequences similar to SARS-CoV-2; I do not believe that the WIV possesses in-house DNA synthesis capabilities (most labs do not), but I would be interested to learn if they do. They would have needed to order synthesized sequences from third-party companies and this would have left a paper trail. The investigation should look into this, but I do not expect anything to be found.
-Samples representing intermediate constructs on the path to the construction of SARS-CoV-2, resident at the WIV.
-Samples from experimental animals demonstrating infection with a similar virus, an intermediate virus, or something
So far there is nothing of the sort in evidence. This is what I mean when I say "there is no evidence."
As a side point, I really don't understand Deigin's obsession with the furin cleavage site. Plenty of coronaviruses have furin cleavage sites. It really isn't too surprising to see one emerge in a poorly-documented species of the virus. Kristian G Andersen takes down several of the arguments that Deigin makes about the furin cleavage site's appearance in a thread here: https://twitter.com/K_G_Andersen/status/1391507230848032772
Anyway, I don't make much of Deigin's arguments. Largely, they are conjecture layered on speculation, with the word "possible" carrying a lot of water for the overall scientific validity of what he tries to say. A lot of things are possible. Show me what is probable, or better yet, show me evidence of what actually happened.